Ovarian hyperstimulation syndrome: cardiac arrest with an unexpected outcome

Abstract

We describe the acute deterioration of a 29-year-old undergoing in vitro fertilisation. Late-onset critical ovarian hyperstimulation syndrome triggered a massive pulmonary embolism and subsequent cardiac arrest. While the prognosis was deemed to be poor, the patient made a full recovery. The potential reasons for this are explored.

Background

Ovarian hyperstimulation syndrome (OHSS) is recognised as one of the most serious complications of assisted reproductive techniques (ART).¹ It either develops in the first 7 days following oocyte retrieval (OR) (early onset) or 9–10 days following OR when implantation has occurred (late onset). Late onset disease tends to be more severe.² While the exact mechanism remains unclear, prolonged exposure of stimulated ovaries to human chorionic gonadotropin (HCG) leads to production of proinflammatory mediators which increase vascular permeability causing shifts of fluid from the intravascular space to the extravascular space.³ This can result in haemoconcentration, electrolyte imbalance, ascites, pleural and/or pericardial effusion, pulmonary oedema, ovarian torsion or cyst rupture, venous thromboembolic events (VTE) and organ failure.^{4 5} The risk of VTEs in OHSS is 1.7% which represents an approximately 100-fold increase from background.⁶

Internationally, OHSS is not a notifiable disease. However, with the growing usage of ART, its incidence is almost certainly increasing. Studies suggest a clinically significant incidence rate of 3%–8%, with up to 30% of in vitro fertilisation (IVF) patients developing mild OHSS.⁷ In the UK, severe and critical OHSS cases must be notified to the Human Fertility and Embryology Authority (HFEA). There were 38 reported cases in the last published HFEA report.⁸ In the UK between 2001 and 2016, there were two deaths registered with OHSS as the cause.⁹

The need for any active management, either inpatient or outpatient depends on the patient’s symptoms and severity of disease. The UK’s Royal College of Obstetricians and Gynaecologists (RCOG) have proposed a severity classification with advice to consider outpatient management in mild and moderate disease and selected severe cases.⁷ OHSS is generally self-limiting with regression of disease spontaneously over 10–14 days as long as prompt and appropriate supportive care is provided.¹⁰

Case presentation

A 29-year-old woman was referred to the assisted conception unit of a London teaching hospital with a 4-year history of secondary subfertility. At the age of 23, she was diagnosed with a delayed miscarriage at 12 weeks’ gestation and 1 year later underwent a laparoscopic salpingectomy for an ectopic pregnancy without complications. Following the miscarriage, investigations in Russia demonstrated ‘abnormal platelet aggregation’. She had no other significant medical history and a body mass index of 23. Standard subfertility investigations performed in the UK for the patient and her partner were normal, including a normal coagulation screen. She had a regular ovulatory cycle and an antral follicle count of 24 and anti-Mullerian hormone (AMH) level of 26.8 pmol/L (55th centile). Due to the duration of subfertility, IVF was recommended. Prior to commencement of IVF, the patient was reviewed in haematology clinic where coagulopathy investigations were normal and therefore discharged.

In late 2018, the patient was stimulated with a short antagonist cycle with minimal Follicle Stimulating Hormone (FSH) dose and oestradiol monitoring. After 10 days of stimulation, oestradiol was 7316 pmol/L and an HCG trigger was administered. Thirty-six hours later, 20 oocytes were retrieved. The patient reported abdominal pain 2 days post egg collection but was otherwise well. On the day of embryo transfer, the pain had settled and bloating was the only symptom recorded. Haemoglobin was 131 g/L and PCV 0.38 /L. Transvaginal ultrasound was unremarkable. The risk of late OHSS was explained and the option to freeze all blastocysts discussed. Assessment for VTE at this stage classified the patient as low risk with no indication for prophylactic heparin. A single day 5 blastocyst was transferred. 9 days after egg collection, the patient attended for assessment due to bloating. There had been no weight gain and transvaginal ultrasound showed bilaterally enlarged cystic ovaries of 10 and 11 cm³ with minimal ascites (figure 1). Based on this, a diagnosis of moderate OHSS was made. As the patient was well, she was managed as an outpatient. The patient was contacted regularly by telephone and reviewed in person twice. She reported ongoing abdominal pain but seemed otherwise stable.

Figure 1.

Ultrasound imagine of right ovary in transverse plane showing increased volume secondary to oedema and multiple cystic areas with internal haemorrhage in keeping with hyperstimulation.

On day 14 post egg collection, the patient was admitted via ambulance with increasing abdominal pain and distension, shortness of breath and a positive pregnancy test. On arrival, observations were within normal parameters. Blood results are shown below in table 1. A bedside ultrasound was performed which revealed ovaries of 14 cm³, gross ascites and bilateral small pleural effusions. At diagnosis of critical OHSS was made and the patient admitted to the gynaecology ward for further investigations and treatment.

Table 1.

Patient blood results

Investigation	Admission	48 hours later	Normal range
Haemoglobin (g/L)	184	143	12–150
White cell count (×106)	18.6	12.3	4.0–11.0
PCV (L/L)	0.56	0.43	0.360–0.47
Platelet (×109)	516	298	150–400
Sodium (mmol/L)	132	140	135–145
Potassium (mmol/L)	4.5	4.3	3.5–5
Creatinine (μmol/L)	102	131	45–84
Albumin (g/L)	38	11	40–52
HCG (IU/mL)	109	16	–

At presentation, 48 hours later and normal ranges.

Over the next 48 hours, symptoms and bloods improved; HCG dropped from 109 to 16 IU/mL suggesting an unsuccessful ET. However, at 06:00, the following morning the patient complained of sudden breathlessness and severe chest and abdominal pain. The patient was in extremis: respiratory rate 35/min; heart rate 135 beats/min, oxygen saturations 70% on 4 L of oxygen and a systolic blood pressure of 75 mm Hg. The patient sustained a pulseless electrical activity cardiorespiratory arrest minutes later. Over the following 35 min, the patient suffered three further cardiac arrests with significant downtime. An Arterial Blood Gas (ABG) was taken (table 2) and a bedside echocardiogram was performed which revealed features of right-sided heart strain in keeping with a PE. Once stabilised, a CT pulmonary angiogram confirmed massive PE. Her partner was counselled at this point that the patient’s chance of survival was in the region of 10% with a very high probability of neurological impairment.

Table 2.

Arterial Blood Gas (ABG) on 100% FiO₂ at time of cardiac arrest

Investigation	100% FiO2	Normal range
pH	6.734	7.35–7.45
pCO2 (kPa)	15.15	4.6–6.4
pO2 (kPa)	15.18	10–14.4
Bicarbonate(mmol/L)	14.9	22–26
Base excess	−20	−2 to +2
Lactate mmoll/L	10.7	0.5–2.2

Differential diagnosis

Abdominal pain and bloating are the most common side effects reported during ovarian stimulation and following egg retrieval.¹ As such, both blood tests and pelvic ultrasound were key in this case in determining the patient was developing OHSS. At the time of OR, a needle is passed from the vagina into the adnexa. As such, infection and injury to surrounding structures including bowel are possible. In this case, the lack of fever, offensive vaginal discharge and gastrointestinal symptoms made this less likely and favoured OHSS.

At the time of acute deterioration, a differential diagnosis of abdominal compartment syndrome with splinting of the diaphragm was considered. However, further deterioration despite opening of the ascitic drain and removal of fluid deemed this to be unlikely.

At the time of oxygen desaturation, in the absence of previous respiratory disease, asthma and tension pneumothorax were both unlikely and unsupported by clinical examination.

When cardiac arrest occurred, it had been preceded by hypoxia. A recorded normal temperature, normal electrolytes and no known structural cardiac disease made these unlikely causes for cardiac arrest. A bedside echo favoured the diagnosis of massive PE and supported management.

Treatment

At the time of diagnosis of moderate OHSS, due to the patient being well, she was managed as an outpatient with daily injections of low molecular weight heparin and advice to stay well hydrated and mobile.

On diagnosing critical OHSS, the patient was admitted to the gynaecology ward for fluid resuscitation with intravenous crystalloids, daily weighing, prophylactic low weight molecular heparin and insertion of an abdominal drain. That evening 2 L of straw-coloured ascites was drained with a plan to reopen the drain the next day.

At the point of deterioration, a further 2 L of ascites was rapidly drained to assist with ventilation. On diagnosis of cardiac arrest, Cardiopulmonary Resuscitation (CPR) was commenced and the patient intubated within 5 min. Return of spontaneous circulations was achieved at minute 10. A streptokinase infusion was commenced for thrombolysis and the patient transferred to intensive care with ongoing automated chest compressions. Inspiratory positive pressure ventilation was continued, venous-arterial extracorpeal membrane oxygenation commenced along with triple vasopressors and hemofiltration.

Outcome and follow-up

Two days following cardiac arrest, ECMO and vasopressors were ceased to good effect. By day 3, the patient was extubated and sedatives weaned. By day 10, dialysis was terminated with normalising renal function. HCG dropped to 0 IU/mL. The patient made a full recovery with no neurological deficit and was discharged on 12 weeks of rivaroxaban anticoagulant. On further discussion, it transpired that the patient had competed as a professional synchronised swimmer for 9 years. Follow-up at 1 and 3 months revealed no physical or psychological sequalae. Haematological investigations were normal, rivaroxaban was ceased and the patient advised to use heparin throughout any subsequent pregnancies. The couple have attempted a further IVF cycle with a frozen embryo which was unsuccessful.

Discussion

This case is of interest for several reasons. First, a complete recovery was unexpected. Mortality rates following cardiac arrest secondary to massive PE are quoted as 95% with neurological deficits being common in survivors.¹¹ However, PE more commonly occur in older patients with comorbidities. In addition to this, the majority of cardiac arrests occur outside of hospital setting. In this case, the fact that the cardiac arrest was witnessed, the situation was managed by a highly skilled multidisciplinary team, thrombolysis was administered rapidly, and the access to ECMO all favoured survival.

Both the European Resuscitation Council and the American Heart Association recommend thrombolysis in PE associated cardiac arrest, despite a lack of good quality evidence.¹² While there is a perceived benefit to its use, it must be remembered that thrombolysis during CPR causes significant haemorrhagic complications in 15% of patients.¹³ Had a member of the resuscitation team not been trained in the use of echocardiography, the acute management of this patient would have been even more challenging.

It is likely that the patient’s years of performing as a professional synchronised swimmer had an impact on the outcome. One study of synchronised swimmers documented the extreme degree of hypoxia tolerated during competitions; alveolar partial pressures of oxygen dropping as low as 3.67 kPa (normal: 12–14 kPa) with no apparent sequalae.¹⁴ Another study demonstrated that despite prolonged apnoea and marked fluctuations in heartbeat, competitors only produced moderate metabolic acidosis suggestive of significant metabolic adaptations from the sport.¹⁵

Cases such as this are rare. As discussed, late-onset OHSS is less common and while it tends to be more severe, progression to critical OHSS is rare. While there have only been two deaths in the UK in the past 15 years from OHSS, both were due to PE.⁹ Prediction of OHSS is difficult. While factors such as young age, high markers of ovarian reserve and polycystic ovarian syndrome may be useful in predicting early OHSS, late OHSS is less predictable.⁷ Despite tailoring the IVF protocol to minimise the risk, and normalisation of oestradiol levels at the end of ovarian stimulation, OHSS still occurred. While much controversy still exists over which triggering medication confers the lowest risk for OHSS development, it is important to remember that late onset OHSS, as in this case, is mediated via endogenous HCG, and as such the choice of trigger has not been shown to alter its incidence.¹⁶

A legal and ethical challenge in cases like this can be when to consider termination of an ongoing viable pregnancy. As previously stated, the vast majority of OHSS cases, even critical, are self-limiting and heparin appears to be non-teratogenic; however, one case report describes progressive thrombosis despite anticoagulation requiring termination of pregnancy.¹⁰ This is supported by the RCOG’s guideline on OHSS management.⁷ In cases where patients lack capacity and next of kin disagrees with termination, or in countries where termination is illegal, this matter should be escalated to the courts urgently.

Finally, counselling of the patient and partner about future fertility is challenging. There is an increased risk of recurrence of OHSS.⁷ Should the couple choose to conceive again, obstetricians and gynaecologists should manage this pregnancy jointly along with haematologists in a unit with expertise in such high-risk cases. In the interim, oestrogen containing contraception should be avoided.

This case reminds us that although rare, critical OHSS can have life changing and threatening consequences. While our patient survived without any sequalae, this is not likely to be the case for all women. The rapid progression in severity of the syndrome in young, fit women should prompt extreme vigilance and a low threshold for thromboprophylaxis at all stages of ART. Multidisciplinary dialogue and care is essential to minimise morbidity and mortality.

Patient’s perspective.

I was going through the IVF treatment, once I completed the hormone therapy and eggs were retrieved, I started to feel that something was not right. My abdomen was bloating and I had pain all the time. I went to see the fertility doctors couple of times, who did some tests and discharged me. One night the pain was unbearable and my husband rushed me to the A&E, where I spent couple of days until I had PE. I was very lucky to be in the hospital, where nurses and doctors could help me in this emergency situation. That was a very traumatic experience and I hope that women, who develop ovarian hyper stimulation would be paid more attention to.

Learning points.

• Ovarian hyperstimulation syndrome (OHSS) is a recognised complication of assisted reproductive techniques and all clinicians should be familiar with it.

• Late-onset OHSS is driven by endogenous human chorionic gonadotropin and while less common, is far more difficult to predict and prevent.

• All patients reporting worsening pain, bloating or shortness of breath following assisted reproductive techniques should be reviewed by a clinician with experience in recognising OHSS.

• OHSS is a hypercoagulable condition and all patients should be assessed for the need for anticoagulation.

• Patients’ premorbid physiological health has significant impact on disease outcomes.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s)